Graduate Computational Algebraic Geometry Seminar
Sonja Petrovic
UIC
The Likelihood Equations
Abstract: In the framework of algebraic statistics, model parameters form (a
part of) an algebraic variety.
Maximum likelihood estimation is concerned with finding those model
parameters that best explain a given sequence of observations; this is
done by maximizing the likelihood function. The defining equations of
the critical points are the likelihood equations, and the number of
complex solutions is the ML degree of the model.
We will discuss some of the current problems with solving these
equations and calculating the ML degree, including a discussion of
best known results.
Thursday September 10, 2009 at 11:00 AM in SEO 612